Research on Attention-Based Multigranular Remote Sensing Image Extreme Object Detection Algorithm
The detection of small objects and high aspect ratio objects is a challenge for remote sensing image object detection. In this paper, we refer to these objects as extreme objects. The Attention-Based Multigranular Extreme Object Detection Network (AMEODNet) is proposed to address these problems. It consists of two parts. The first part is a multi-granularity information fusion module. The second part is an extreme object extraction section. The proposed method is tested on the NWPU VHR-10 datasets. The results are compared with mainstream object detection algorithms to verify the effectiveness of the proposed method.
Authors
- Weihua Xu (ORCID: https://orcid.org/0000-0001-7202-2573)
- Meng Li (ORCID: https://orcid.org/0000-0002-1543-5029)
- Eric C.C. Tsang (ORCID: https://orcid.org/0000-0002-3734-7273)
- Qiang He (ORCID: https://orcid.org/0000-0002-3505-7010)
Institutions
- Macau University of Science and Technology (MO)
- Southwest University (CN)
- Beijing University of Civil Engineering and Architecture (CN)
Publication Details
- Published
- 2024-09-20
- DOI
- https://doi.org/10.1109/icmlc63072.2024.10935211
- Primary Topic
- Remote Sensing and Land Use
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Fundo para o Desenvolvimento das Ciências e da Tecnologia